Extract/partial-mash differences

I can't figure out why the original gravity estimates differ so much between Extract and Partial-Mash recipe types. For example, if you make a 5-gallon recipe with nothing but 7 lbs of pale LME, BeerSmith estimates an OG of 1.046 if you call it an Extract recipe, but 1.050 if you say it's a Partial Mash. If you add a pound of Crystal 40L (which doesn't need mashing) the difference is even bigger - Extract calls it 1.047, but Partial Mash says 1.055.

The OG difference between mashing and steeping crystal malt should be negligible, and of course there should be no difference where the extract is concerned.

What am I missing here? What are the assumptions that go into each type?

I'm really sorry you have this level of confidence, because it's based on false assumptions.

All extract efficiencies in brewing are based on sucrose, which yields 1.046 as 1 lb of sugar to make 1 gallon of solution. Basically, it's 100% efficient. Grain simply is not. With extract, everything that is soluble is already a part of it. Dry is higher gravity per lb because it contains less water.

Grains are, at best, 82% soluble compared to sugar. Often less. When you change the recipe from extract to partial mash (or all grain), enzymes and a greater yield are assumed. Any variation is due to the rounding error of specific gravity.

Crystal/dextrin malts are NOT 100% "non fermentable" nor are they 100% efficient. What they contribute to a wort is directly related to the brewer's methods. Even a 100% crystal malt wort is at least 50% fermentable. The fermentability goes up when the dextrin malt is exposed to base malt; becoming more fermentable with contact time to enzymes. Mashing dextrin malts converts the residual starch and increases the gravity compared to simple steeping.

Thanks, but I didn't ask anything at all about fermentability.

Crystal malt has little residual starch; it's mostly converted to sugars in the malting process. It won't make much difference whether you mash it or steep it. John Palmer's experiments indicate Crystal 40L will deliver 22 ppppg when steeped in 1 gal water/lb. BeerSmith indicates 26 ppppg when mashed at 75% efficiency; fine, I'll give you another 4 ppppg for mashing.

But when you set it for extract (steeping), it estimates only 3 or 4 ppppg. No way that's correct, and it isn't "rounding error". Again, this is original gravity; it has NOTHING to do with fermentability.

Regardless of all that, it doesn't explain why you get very different results when using only extract. How can the OG of a 100% extract beer depend on whether you're doing "extract brewing" or "partial mashing"? If there is no grain, there is no difference. The gravity contribution of extract can't depend on the method.

First, I probably shouldn't post after a GABF victory party...  : Second, yes, I misunderstood your question. My bad, x2.

With settings of:

• extract
• no trub loss

Using your model of 7 lb LME for 5 gallons, I get 1.050 as extract, partial and full mash versions. No difference.

As extract only, errors begin to occur when adding in trub loss for an extract batch. The more the trub loss, the wider the error. Clearly it's a math error because it's predicting more gravity from the same fully soluble material in a greater amount of liquid.

This seems to be due to how BeerSmith predicts mash efficiency increases with increasing trub loss. I'm observing the gravity rise even with decreased brewhouse efficiency, which is how the trub loss should be addressed in BeerSmith. That seems like a bug, to me.

Moving on to grain.

Going back to the baseline.

• 100% efficiency
• no trub loss
• mash ratio of 1.25:1
• resting at 148F

In partial and full mash versions, the grain models correctly. That is, that 1 lb in 1 gallon equals the full potential of the grain. Adding 0.5 gal trub loss to 5 gallons, the gravity falls by 9%, which is also correctly modeled (5 / 5.5 = 90.9%).

In extract, the grains become worth about 22% of their potential. It's about half of what Palmer states and what this experiment found: http://www.homebrewtalk.com/f128/testing-fermentability-crystal-malt-208361/index11.html

Closer contribution numbers can be created by boosting the grain potential to 100%. Simply double click the grain and change the yield field in the extract recipe. This leaves the grain database intact for future all grain batches.

However, in this case, fermentability is a factor because steeped grain attenuation is so low. The base extract is much more fermentable. If steeped grain were given a higher potential and the same fermentability of 75 - 80% (which that link shows is not true), real world extract beers would finish much higher than predicted. So, the low contribution number seems to be offset by the statistical reality of what steeped specialty grains do in a beer.

Kudos to you sir; for a post-party post, the quality of your spelling and grammar is quite impressive ;-)

I didn't think to toy with the trub loss settings. There does seem to be something going on; I agree that there must be a math error in there somewhere. I'll play with it and see if I can nail it down further.

There may be more than one problem though - when I play with a 100% crystal malt recipe, even with zero trub loss, to me the results are pretty unexpected. I'm getting more like 12% of potential. I need to play with some of the other variables.

I'm a little surprised nobody has complained about this before, but I suppose BeerSmith might not be used often for extract brewing. I came across this when trying to convert one of my all-grain recipes to extract for a friend - the numbers just weren't making sense.